Rocket Learning Scales its Digital Classroom Solutions by 50% to Support 2 Million Users on AWS

In India, 35 million low-income children lack quality early childhood education. Rocket Learning aims to change this. The growing educational technology nonprofit fosters digital communities of parents and teachers to provide relevant content to students in low-income regions. Rocket Learning imparts foundational learning to 1.75 million children through 200,000 digital classrooms in eight states.

The organization continually develops new solutions to enhance learning in its communities. Its current focus is on the integration of advanced artificial intelligence (AI) to develop an AI and machine learning (ML) coach and model aimed at transforming the behavior of low-income parents and teachers. “This scalable solution has the potential to impact tens of millions of lives,” says Vishal Sunil, chief technology officer at Rocket Learning. Expanding the solution was challenging, however, because the organization’s existing workloads—including both internal solutions and customer-facing educational content applications—were hosted in a cloud environment that restricted growth. “As we grew to two million users and 100 million interactions, we faced difficulties in scaling up our platform and servers,” says Sunil.

Additionally, Rocket Learning experienced challenges with the availability and reliability of its IT environment. Developers depended on manual application deployment processes, resulting in longer lead times to deliver new features and updates to customers. "We aimed to modernize our environment to maintain agility and adaptability in a rapidly evolving landscape," Sunil says. "Ensuring our applications remained competitive and efficient moving forward was crucial."

Rocket Learning engaged CloudThat Technologies, an AWS Partner, to migrate its cloud environment to AWS. The migration aimed to boost scalability, enhance performance, and achieve cost efficiency.

“CloudThat significantly aided our AWS migration,” says Sunil. “With a dedicated team at our disposal, the transition was seamless. We easily moved our database and storage accounts, and their recommendations strengthened our architecture, making it more robust and efficient on AWS.” Rocket Learning decided to adopt a microservices-based architecture to increase the flexibility and scalability of its applications and develop and roll out specific functions independently.

CloudThat used AWS Database Migration Service (DMS) to migrate Rocket Learning’s databases to Amazon Relational Database Service (Amazon RDS) and implemented Amazon Elastic Kubernetes Service (Amazon EKS) to support its microservices approach. It also deployed AWS Auto Scaling to automatically adjust compute capacity on demand. Additionally, CloudThat helped Rocket Learning automate deployments with continuous integration/continuous deployment best practices and ensured all applications were aligned with the AWS Well-Architected Framework.

To boost the performance of Rocket Learning’s applications, CloudThat deployed Amazon Elastic Compute Cloud (Amazon EC2) instances powered by AWS Graviton Processors. Rocket Learning subsequently adopted Amazon EC2 Reserved Instances to reduce overall compute costs for long-running applications. As a result, Rocket Learning achieved cloud cost savings of 20–25 percent. Finally, the company implemented AWS Identity and Access Management (IAM) to enhance security.